Force delete support for ASGs

[acasas]

This patch should allow users to force-delete ASGs. It follows a similar approach to what was already implemented for clusters, using the GroupDeleteOperation class.

[joesondow]

It's helpful to include the Jira ID at the start of the pull request name, and at the start of one of the commits within the pull request.

[cloudbees-pull-request-builder]

asgard-pull-requests #70 SUCCESS This pull request looks good

[joesondow]

Almost all of the code of the method deserves to be deleted and replaced with something more similar to what Cluster.delete does. The booleans and instance count protection and the try-catch blocks were only present because we were doing a synchronous Amazon call with a lot of edge cases to handle.

[joesondow]

This actually wouldn't be an improvement for the user, since the purpose of using Force Delete is that it's possible to delete an ASG that still contains some instances, whether the instances are real or phantoms. There was an internal email this afternoon about this very problem.

[cloudbees-pull-request-builder]

asgard-pull-requests #71 SUCCESS This pull request looks good

[cloudbees-pull-request-builder]

asgard-pull-requests #72 SUCCESS This pull request looks good

[claymccoy]

@joesondow Is there value in the extra work that GroupDeleteOperation is doing here? I'm asking not really for this issue, but to see if it should be in the new auto deploy workflow.

[cloudbees-pull-request-builder]

asgard-pull-requests #73 SUCCESS This pull request looks good